US4786820A - Apparatus for parallel-processing image data - Google Patents

Apparatus for parallel-processing image data Download PDF

Info

Publication number
US4786820A
US4786820A US07/147,091 US14709188A US4786820A US 4786820 A US4786820 A US 4786820A US 14709188 A US14709188 A US 14709188A US 4786820 A US4786820 A US 4786820A
Authority
US
United States
Prior art keywords
image
processing
digital
signal
input
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Expired - Lifetime
Application number
US07/147,091
Other languages
English (en)
Inventor
Yoshitaka Ogino
Hiroshi Tanioka
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Canon Inc
Original Assignee
Canon Inc
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Canon Inc filed Critical Canon Inc
Application granted granted Critical
Publication of US4786820A publication Critical patent/US4786820A/en
Anticipated expiration legal-status Critical
Expired - Lifetime legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T1/00—General purpose image data processing
    • G06T1/20—Processor architectures; Processor configuration, e.g. pipelining
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N1/00—Scanning, transmission or reproduction of documents or the like, e.g. facsimile transmission; Details thereof
    • H04N1/40—Picture signal circuits
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N1/00—Scanning, transmission or reproduction of documents or the like, e.g. facsimile transmission; Details thereof
    • H04N1/40—Picture signal circuits
    • H04N1/403—Discrimination between the two tones in the picture signal of a two-tone original

Definitions

  • This invention relates to an image processing apparatus such as a copying apparatus or a facsimile apparatus, and in particular to an image processing apparatus which effects parallel-processing by a plurality of image processing means.
  • an image has been raster-scanned and read by a CCD or the like and has been subjected to image processing such as a binarizing process or tone processing.
  • image processing such as a binarizing process or tone processing.
  • it has been proposed as the processing of this type to recognize the tone of the image on real time and selectively change over a processing suitable for the image tone to thereby reproduce the image.
  • the processing algorism is complicated to discriminate accurately and accordingly, taking the scale of the circuit or the processing speed into consideration where the apparatus is constituted by hardware, the apparatus is not yet practical to such a degree that it is applicable to a versatile copying apparatus.
  • an object of the present invention to provide an image processing apparatus which is made high in image processing speed and compact in size by parallel-processing the image input information in each block.
  • an object of the present invention to provide an image processing apparatus in which, by inputting the data of a pixel to be processed by another image processing means adjacent to an image processing means, the image processing between the image processing means is endowed with conformity.
  • FIG. 1 is a block diagram showing an embodiment of the present invention and illustrating the wiring of each construction.
  • FIG. 2 is a block diagram showing the internal construction of the CPU in image processing means.
  • FIG. 3A shows the memory layout of the data memory in the CPU and the image signal inputting and processing system.
  • FIG. 3B shows the layout of the internal registers in the CPU.
  • FIG. 3C shows a method of image tone recognition in the data memory.
  • FIG. 4 is a flow chart schematically showing the image processing according to the method of an embodiment of the present invention.
  • FIG. 5A is a main flow chart showing the image processing according to an embodiment of the present invention.
  • FIGS. 5B, 5C, 5D, 5E, 5F, 5G, 5H, 5I and 5J are flow charts showing the details of image processing.
  • FIGS. 6A, 6B, 6C and 6D show methods of wiring on a base plate.
  • FIGS. 7A, 7B, 7C, 7D, 7E, 7F, 7G and 7H show threshold matrices for binarization.
  • FIGS. 8 and 9 show another method of coupling between CPUs.
  • FIG. 10 shows the input system when image information is input by another method of coupling.
  • boundary cell information is overlappingly introduced into adjacent CPUs and image processing is effected.
  • FIG. 1 is a block diagram of an embodiment of the image processing apparatus of the present invention.
  • the components of each block are a sensor cell, a sampling and holding circuit, the CPU, a shift register, etc. Each component will now be described.
  • E1, E2, . . . , E41 designate a group of sensor cells arranged in the form of a line.
  • each sensor cell is of the thin film type which comprises amorphous silicon and has a characteristic that the resistance value thereof varies in conformity with the quantity of light. If an area provided with a light-intercepting mask is connected in series with each cell area and a power source is connected to the opposite ends to thereby detect any potential change at the junction between two areas, the quantity of light received by the cell area can be known.
  • the construction of the sensors can likewise by realized by the use of the type which produces an electromotive force in accordance with the quantity of light.
  • the outputs of the sensors of this type can be connected at one-to-one rate to the A/D input ports of CPU1, CPU2, CPU3, . . . through sampling and holding circuits SH0, SH3, . . . (which are not necessary in principle).
  • an analog signal is serially transferred over a long distance and therefore, in spite of the high-speed responsiveness of the cell itself, high-speed reading has been impossible, but by using, for example, only the photodiode portion of the one-to-one magnification type CCD sensor of the present embodiment, the time required for the serial transfer becomes unnecessary and high-speed reading becomes possible.
  • An original is illuminated by a linear light source such as a fluorescent lamp or an LED, the image of the original is formed on a sensor row by a one-to-one imaging optical system such as Celfoc, and the sensors and the original are sub-scanned by conventional scanning means for moving the sensors and the original relative to each other in a direction perpendicular to the direction of arrangement of cells.
  • a signal of a predetermined period put out from said scanning means for example, in the present embodiment, the signal LA of FIG. 1 input at a period of 400 ⁇ sec.
  • divided image information can be read one line by one line by the CPU at a time.
  • sampling and holding circuits (SH), CPUs and shift registers (SR) may respectively be of the same type and therefore, a number of them can be disposed on the same base plate, and this contributes to the compactness of the apparatus. Further, the distance from the sensor cell to the A/D conversion becomes shortest and therefore, image reading of a good S/N ratio and high reliability becomes possible.
  • the image tone recognition/image processing effected by each CPU during one input cycle is for only eight pixels as will later be described, and the burden allotted to a CPU for the image processing is small and therefore, even considerably complicated processing (which will later be described) becomes possible within a short time and further, real time image processing also becomes possible by microcomputerizing the CPUs and executing microcodes.
  • image tone discrimination is effected and a binarizing process is effected during said cycle.
  • the binarized image signals are put out from the output ports 01-08 (FIGS. 1 and 2) of the CPUs, are parallel-latched by concatenately coupled (series-connected) shift registers SR1, SR2, . . . (FIG. 1) and are successively put out is serial binary signals to a printer or the like by a high-speed clock CLK.
  • the information corresponding to one line is divided by eight pixel units and one unit thereof is divisionally allotted to each CPU and therefore, if continuous two-dimensional processing of the image signal is effected, there may arise non-conformity of image processing between adjacent CPUs as will hereinafter be described.
  • CPU 1 further requires the cell information of cells E7, E9, etc. in order to process the cell information of cell E8 in the above-described manner.
  • the cell information of cell E9 in charge of CPU 2 becomes necessary.
  • the information of cell E8 becomes necessary.
  • allotment of pixels is effected in the following manner and the pixel input by each CPU is determined. If the size of the mask applied to image tone recognition or image processing is (2n+1) ⁇ (2n+1), each CPU is designed to further input n pieces of cell information out of the cell information in charge of two adjacent CPUs. Also, dividing the cell information of one line into how many blocks is determined by how many CPUs take charge of one line. So, in the CPUs of the present embodiment shown in FIGS.
  • the size of the mask (matrix) for effecting image tone recognition is 9 ⁇ 9 and the input capacity of each CPU is, for example, sixteen pixels, and therefore, the output of each sensor cell is divided into eight pixels and each CPU is caused to take charge of this 8-pixel cell information and further, of this 8-pixel information, the first 4-pixel cell information and the last 4-pixel cell information are also input to two adjacent CPUs.
  • 4-pixel cell information is input from the two adjacent CPUs also to said CPU.
  • each CPU 8 pixels out of the pixels to be processed by that CPU and 4 pixels from each of two adjacent CPUs are further input.
  • the pixel information from E8, E7, E6 and E5 is input, each likewise, for the rightmost pixel E16, the pixel information from E17, E18, E19 and E20 is input. That is, 16 pixels of E5-E20 are input to CPU 2 to process 8 pixels of E9-E16.
  • the two-dimensional image processing based on 9 ⁇ 9 image information encircling the desired pixel becomes possible in principle and further, highly accurate processing becomes possible also for the pixel positioned near and between two adjacent CPUs without the continuity thereof being lost.
  • connection shown in FIG. 1 is an example of the case where the density of cell arrangement is equal to the density of CPU input port terminals, and where as described above, connection is overlappingly made to each cell (where the density of CPU input port terminals is higher than the pattern density), input terminals are disposed on the opposing sides of adjacent CPU chips by the use of a chip carrier or the like, and a pattern is formed so that the opposing terminals of the two adjacent chips are "staggeredly" arranged, whereby connection is constructed by the high-density actual mounting by three surfaces.
  • the density of CPU input port terminals is low as compared with the density of cell arrangement, the connection by only one surface of each chip is of course possible.
  • FIGS. 6A to 6D there are connecting methods shown in FIGS. 6A to 6D.
  • the connecting method of FIG. 6A is the same as the connecting method of FIG. 1.
  • FIGS. 6A to 6D attention is paid to the input wiring from the sampling and holding circuit (SH) to CPUs. That is, FIGS. 6A and 6D show the wiring when use is made of a QIP (Quad Inline Package) type IC package or chip carrier.
  • dots at the points of intersection between the IC package and the wiring are input terminals and show that input is effected therefrom into IC chips.
  • dotted lines show that the wiring on the print wiring base plate is hidden by the IC package or the like. For example, if a number of CPUs are actually mounted on a ceramic multichip base plate having a great thickness as shown in FIGS. 6A to 6D, the wiring will not intersect and actual mounting of high density will become possible.
  • FIG. 2 is a block diagram showing the internal construction of the 1-chip microcomputer (hereinafter referred to as CPU) of the image processing apparatus.
  • the portion encircled by a dotted line is the CPU which contains a successive comparison type A/D converter of 16 channels in an 8-bit microprocessor.
  • These 16 analog inputs AN1-AN16 are multiplexed in the chip by two multiplexers, an input is selected in accordance with the setting of an A/D channel mode register, and this selected analog input is digitalized by two A/D converters having the following construction.
  • Each A/D converter respectively comprises a sampling and holding circuit SH4 (or SH5), a voltage comparator 3 (or 4), a tap decoder 3 (or 4) and an SAR 3 (or 4).
  • the voltage comparator 3 compares the differences between said sampled and held analog input and the voltage taps of the serial resistance string in the tap decoder 3 (or 4).
  • This serial resistance string is connected between a terminal (not shown) for receiving an external reference voltage Vref for A/D and a ground terminal for A/D (not shown), and is constituted by 256 equivalent resistors for making the voltage between the two terminals into 256 equivalent voltage steps.
  • the voltage taps of the serial resistance string are selectively driven by an 8-bit shift register SAR 3 (or 4).
  • the SAR 3 (or 4) sets by 1 bit each from the most significant bit (MSB) of the SAR 3 (or 4) until the output of the voltage tap of the serial resistance string becomes coincident with the analog input (that is, until the output of the comparator is inverted).
  • the (MAB) of the SAR 3 (or 4) is set, and 1/2 Vref of the voltage tap of the serial resistance string is selected and compared with the analog input. If the analog input is greater than 1/2 Vref, the MSB of the SAR 3 (or 4) is left as it is set, and if the analog input is smaller than 1/2 Vref, the MSB is reset. Subsequently, the bit next to the MSB is set and the voltage tap of the serial resistance string is rendered into 3/4 Vref or 1/4 Vref, which is compared with the analog input. The respective bits from the most significant bit to the least significant bit are compared with the voltage value selected by the SAR 3 (or 4) while being successively changed in this manner.
  • the SAR 3 (or 4) holds the digital value of the analog input and the result is latch-input to registers CR1-CR4 (CR5-CR8).
  • the A/D channel mode register by its being set, can select analog inputs AN1-AN4 (and AN9-AN12) or AN5-AN8 (and AN13-AN16) at a 4-channel unit, and assuming that, for example, AN1-AN4 (AN9-AN12) have been selected, the analog inputs are selected and A/D-converted in the order of AN1 (AN9) ⁇ AN2 (AN10) ⁇ AN3 (AN11) ⁇ AN4 (AN12).
  • the A/D-converted values of the respective inputs are stored in the order of CR1 (CR5) ⁇ CR2 (CR6) ⁇ CR3 (CR7) ⁇ CR4 (CR8).
  • A/D interrupt an internal interrupt requirement (hereinafter referred to as A/D interrupt) takes place.
  • a priority order is provided between the interrupt on the CR1-CR4 side and the interrupt on the CR5-CR8 side, and even if two A/D interrupt requirements take place at a time, they will be processed in accordance with the priority order.
  • the degree of priority is higher on the CR1-CR4 side.
  • the interrupt procedure in the control flows of FIGS. 5A to 5J which will later be described is such that the A/D conversion of the channels AN1-AN4 and AN9-AN12 is first started.
  • Each A/D converter effects A/D conversion and, when there is A/D interrupt, processing of the channels AN1-AN4 stored in CR1-CR4 is effected and the A/D interrupt on the CR5-CR8 side is waited for.
  • processing of the channels AN9-AN12 stored in CR5-CR8 is effected, whereafter the A/D conversion of AN5-AN8 and AN13-AN16 is started and processing similar to what has been described above is effected.
  • the A/D conversions of the four pixels of the group of AN9-AN16 can be caused to overlap one another to thereby improve the total throughput during the time that the four pixels of the group of AN1-AN8 are image-processed.
  • the components are an ALU (Arithmetic Logic Unit) for effecting operations, input ports I 0 -I 7 and output ports O 1 -O 8 , for effecting the input-output interface from the outside (signals MOD1-8 are input to the input ports), a program memory storing therein a control program in which the control procedures processing procedures as shown in FIGS.
  • ALU Arimetic Logic Unit
  • a program counter for holding an instruction execution address
  • an instruction register for holding an instruction code indicated by the program counter
  • a high-speed internal register (the details of which will later be described) used when the microprocessor executes the instruction
  • a data memory which is a main memory capable of reading and writing
  • a PSW Program Status Word
  • 2K bytes is prepared as the capacity of the program memory and 256 bytes is prepared as the capacity of the data memory (RAM).
  • RAM data memory
  • AN1-AN16 As the terminals of the CPU, there are AN1-AN16, I 0 -I 7 (input ports), O 1 -O 8 (output ports), and an INT terminal which is an external interrupt input, and as unshown terminals, there are power source terminals Vcc, Vss, AVcc (power source for A/D) AVss (GND for A/D), Vref (reference voltage input for A/D), RESET terminal, START signal terminal, STOP signal terminal, and X1, X2 (such as crystal connection terminals for oscillating contained clocks).
  • the interrupt signal from outside the CPU for example, LA (Line Advance) signal obtained from a horizontal synchronizing signal in the sub scan direction of the input system, is input to the INT terminal.
  • LA Line Advance
  • FIG. 3A shows the memory layout of image signals, etc. after A/D conversion stored in the data memory.
  • X and Y are the X counter and Y counter in the register within the CPU which will later be described.
  • and b(Y) ⁇
  • A(X,Y) in FIG. 3A is a memory map obtained when the control procedures of FIGS. 5A-5J have been executed in the state of the sensor cells, the sampling and holding circuit, the CPU and their wiring shown in FIG. 2.
  • this CPU binarizes the image information of the positions of coordinates (5, 5), (6, 5),-(12, 5), i.e., the image information of the sensor cells E9-E16, while taking the image information E5-E8, E17-E20 of adjacent sensor arrays into consideration.
  • a plurality of such CPUs effect binarization while effecting image tone recognition in parallel and, when all the CPUs terminate the binarization of 1 line, the sensor cell advances by 1 line in the sub scan direction.
  • the image information in the 16 ⁇ 9 memory map is shifted by 1 line in the sub scan direction (the direction of arrow).
  • Newly A/D-converted image information is stored at the coordinates positions (1, 9)-(16, 9) of an empty line 9. Then, the aforementioned image tone recognition, binarization and line shift are repeated.
  • the CPUs can accomplish the image tone recognition and binarization of the image information from E9-E16 two-dimensionally continuously over the main scan direction and the sub scan direction.
  • W may be varied from 0 to 7.
  • FIG. 3B shows the layout of the internal register.
  • the internal register there are various registers of 8-bit width, counters and 1-bit flags.
  • RES (result) register holds the result obtained by binarizing eight pixels (5, 5), (6, 5)-(12, 5) on the basis of T (threshold) register.
  • Max register holds the maximum density value in the 9 ⁇ 9 matrix
  • min register holds the minimum density value in the 9 ⁇ 9 matrix.
  • BLC counter is a counter for selecting a threshold value from a threshold matrix (8 ⁇ 8) which will later be described, and assumes a value of 0-63.
  • C register is a register for placing the aforementioned result of the binarization of eight pixels (5, 5), (6, 5),-(12, 5) into a predetermined position in the RES register.
  • F/FL FL (Flag for First Line completed) and F/OUT (Flag for Output) are flags for the image processing apparatus to distinguish the initial start time from the others in the control of the processing. That is, F/FL flag is for the distinction of the fact that the density difference between adjoining pixels in the sub scan direction cannot be found for the first line during the initial start, and on the other hand, F/OUT flag is for the distinction of the fact that the image tone recognition of the 9 ⁇ 9 matrix cannot be accomplished until all data are stored in the 16 ⁇ 9 memory map, but after that, the image tone recognition can be accomplished each time 1 line is A/D-converted.
  • FIG. 4 is a schematic flow chart at a macro level of the processing procedure in which the CPUs effect two-dimensional image tone recognition, select a threshold value and binarize it. Description will first be made of this schematic flow chart.
  • AN1-AN16 of 1 line are A/D-converted and stored in the CR register, whereafter at step 11, the sum of the absolute values of the density differences between adjoining pixels in the main scan direction is stored in b(Y).
  • F/OUT flag is a flag for effecting image tone recognition for each line once the image information has been stored in the 16 ⁇ 9 memory map, because the 9 ⁇ 9 matrix of the present embodiment effects image tone recognition while shifting by a pixel each in the main scan direction and also effects image tone recognition while shifting by a line each in the sub scan direction.
  • step 13 the procedure proceeds to step 13, at which the sum of the density differences between adjoining pixels in the sub scan direction (the pixels of the preceding line) is integrated in a(X), and at step 14, the contents of the CR register (i.e., the image information of the A/D-converted line) are stored in the 16 ⁇ 9 memory map.
  • the sensor cell advances by 1 line in the sub scan direction (step 15), and steps 10 to 16 are repeated until the storage for 9 lines is completed.
  • F/OUT flag is set (step 17).
  • the 9 ⁇ 9 matrix is set from the left end of the 16 ⁇ 9 memory map (step 18), and the sum of the density differences between adjoining pixels in the main and sub scan directions is integrated in a register S, not shown (at this time, the sum a(X) of the density differences between adjoining pixels in the sub scan direction is utilized) (step 19), and at steps 20 and 21, the maximum and minimum density values in the 9 ⁇ 9 matrix are found and stored in the max and min registers, respectively.
  • image tone recognition is effected on the basis of a formula. Where the interior of the 9 ⁇ 9 matrix is of the same density value, S/(max - min) is 0/0, but is regarded as O.
  • a threshold value is selected in accordance with the result of the image tone recognition, and at that threshold value, the central pixel of 9 ⁇ 9 is binarized and stored in the RES register (step 24).
  • the 9 ⁇ 9 matrix is shifted in the main scan direction (step 25), and said operation is repeated until it is effected on the central pixel of 8 pixels (step 26).
  • step 27 the preparation for shifting the memory map 1 line by 1 line is made at step 27. That is, the density difference between the adjoining pixels of line 1 and line 2 is subtracted from a(X) (step 27). Then, the memory map is carried over by 1 line (step 28). The content of the RES register (the binary data of 8 pixels in charge of the CPU) is put out from the output port (step 29), the RES register is cleared (step 30), and the procedure returns to step 10.
  • the RES register the binary data of 8 pixels in charge of the CPU
  • step 31 the density difference between the image information thereof and line 8 is found and a(X) is renewed (step 31), and the image information concerned is stored in the 16 ⁇ 9 memory map of line 9 (step 32), and thereafter the aforedescribed image tone recognition and binarization are repeated.
  • FIG. 5A shows the main flow of the processing program.
  • the CPU initializes the input/output port, the data memory, the A/D channel register and the interrupt, and at step 41, it waits for the START signal from outside.
  • This START signal is the signal from other control means or the like for generalizing the entire CPU and provides the starting of the image input.
  • the CPU Upon reception of this START signal, the CPU allows the interrupt at step 42, and halts the processing at step 43.
  • This HALT state is released only by the program counter being incremented by a return instruction after the processing of the interrupts, or by RESET (signal or re-closing of the main switch). Accordingly, after the processing of the external interrupt (LA) or the A/D interrupt, the main flow proceeds to step 44.
  • LA external interrupt
  • A/D interrupt the main flow proceeds to step 44.
  • step 44 whether the processing should be completed is judged and if there is no STOP signal, the main flow returns to step 43, at which the CPU again assumes the HALT state, and if there is STOP signal, the main flow proceeds to step 45.
  • This STOP signal like the aforementioned START signal, is supplied from other control means.
  • step 45 the external interrupt (LA) and the A/D interrupt are inhibited, whereafter the main flow returns to step 40, and the above-described processing is repeated.
  • FIG. 5B shows the flow of interrupt waiting and interrupt processing during the HALT. Accordingly, the interrupt waiting after the aforementioned external interrupt allowance or the A/D interrupt allowance to be described is effected at step 50 of FIG. 5B.
  • an interrupt is waited for. If there is an interrupt, whether that interrupt is by the external interrupt or by the A/D interrupt (there are four kinds of this interrupt as will later be described) is examined, and the interrupt address is decided from the interrupt vector (step 52).
  • step 54 the return address is decided (in the present embodiment, step 44 is the return address), and at step 56, the various registers in the microprocessor and the return address are saved in PSW register so that the initial state can be restored even if the return is effected, and the flow jumps to said interrupt address (which, in the present embodiment, is step 60, 70, 100, 130 or 160 depending on the kind of the interrupt) (step 58).
  • step 50 the flow jumps to an address shown in that interrupt vector address (in the present embodiment, step 60).
  • step 60 flag F/OUT is examined (this flag is a flag for indicating whether the first 9 lines after the aforementioned initial start have all been stored in the data map).
  • this flag is set at least indicates that the density data of all pixels are stored in the data map and the image processing up to the preceding line (in the present embodiment, the binarization of 8 pixels) is also completed and therefore, the flow proceeds to step 62, at which the register RES in which the image-processed 8-bit output data are stored is put out to the output port, whereafter the register RES is cleared and the flow proceeds to step 64.
  • the flag F/OUT is not set, it means that all data are not yet stored in the data map, and therefore the register RES is not put out and the flow directly proceeds to step 64.
  • step 64 data for selecting channels An1-An4 and An9-An12 is imparted to the A/D channel register for setting the A/D conversion mode, and at step 66, the inhibition of the external interrupt (LA signal) and the A/D interrupt is released, and at step 68, an A/D start instruction for effecting A/D conversion in accordance with the content of the A/D channel register is supplied to the A/D converter, and the flow returns to the return address to which the external interrupt is applied, and waits for the A/D interrupt.
  • LA signal external interrupt
  • step 68 an A/D start instruction for effecting A/D conversion in accordance with the content of the A/D channel register is supplied to the A/D converter, and the flow returns to the return address to which the external interrupt is applied, and waits for the A/D interrupt.
  • FIGS. 5D to 5J show A/D interrupt processing routines.
  • step 50 When the A/D interrupt takes place at step 50, the flow jumps to the address shown in the vector address, as previously described.
  • step 70 When the interrupt of An1-An4 takes place, the flow jumps to step 70, and at steps 70 to 74, the absolute values of the differences between the registers CR1-CR4 in which the results of the A/D conversions of four channels An1-An4 are stored are found and stored in the data memory b(y). Subsequently, at step 76, the state of flag F/FL is examined, and this flag is not set for the first line and therefore, the flow proceeds to step 78. At step 78, the contents of the registers CR1-CR4 which store therein the pixel density of the current line are intactly stored into A(1, 1)-A(4, 1). It is because there is no preceding line relative to the first line that said contents are intactly stored. The flow then returns to step 50 and waits for the interrupt of the next An9-An12.
  • the next interrupt takes place from the channels 5-8.
  • the absolute values of the density differences in the main scan direction are accumulated in b(y) at steps 130 to 134 as at the aforedescribed steps 100 to 104.
  • the flags are examined and the processing of the first line is not yet completed and therefore, the flow proceeds to step 138, at which the pixel densities of An5-An8 stored in CR1-CR4 are stored in A(5, 1)-A(8, 1), and the flow again returns to step 50 and waits for the interrupt from the channels 13-6.
  • the absolute values of the pixel density differences in the main scan direction are accumulated at steps 160 to 164, and at step 168, the pixel density of the first line is stored in A(13, 1)-A(16, 1).
  • the flow then proceeds to step 200 of FIG. 5H.
  • the absolute value of the density difference of the first line in the main scan direction is stored in b(y) and the pixel densities of An1-An16 are stored in A(1, 1)-A(16, 1). Since the other CPUs are also effecting the above-described operation at the same time, the pixel data over the entire 1 line is stored.
  • step 200 y counter counts up by 1 in preparation for the next line.
  • step 204 the binarization of the central pixel is effected while the image tone recognition in the 16 ⁇ 9 memory is effected, but as a rule, for the tenth and succeeding lines, the image tone recognition and the binarization of the central pixel are effected at each 1 line and therefore, even if y ⁇ 10 at step 202, the flow proceeds to step 212 and subsequent steps, and for this reason, step 204 is provided.
  • step 204 the flag is examined, whereafter the flow proceeds to step 206, at which flag F/FL indicating that the storage of the data of the pixel of the first line has been completed is set, and the flow returns to step 50 to wait for the next external (LA signal) interrupt.
  • step 70 When the interrupt from the channels 1-4 takes place, the flow jumps to step 70, and steps 70 to 74 are repeated, whereafter the flow proceeds to step 76. Since flag F/FL is now set, the flow proceeds to step 80, at which the sum of the absolute values of the differences between CR1-CR4 storing therein the pixel density of the current line (the second line) and the pixel densities A(1, y-1)-A(4, y-1) of the preceding line (y-1) stored in the data memory, i.e., the absolute values of the density differences in the sub scan direction, is accumulated in a(1)-a(4).
  • step 88 flag F/OUT is examined and the processing of 9 lines is not yet completed (NO) and therefore, the flow proceeds to step 90, at which the s of the pixels of An1-An8 are stored in A(1, y)-A(4, y) corresponding to the current line indicated by the counter y, and then the flow returns to step 50 and waits for the interrupt from the next An9-An12.
  • the process of accumulating the sum of the absolute values of the density differences in the main scan direction is effected at steps 100 to 104, and further, at step 106 and steps 110 to 116, the accumulation of the sum of theabsolute values of the density differences in the sub scan direction is effected, and at steps 118 and 120, the contents of CR5-CR8 in which the pixel density of the current line is stored are stored in A(9, y)-A(12, y) indicated by counter y.
  • step 122 the A/D channel register is set to effect the A/D conversion of An5-An8 and An13-An16, and an A/D conversion start instruction for effecting the A/D conversion is executed in accordance with the content of this A/D channel register, and the flow returns to step 50 to wait for the interrupt from the channels (5-8).
  • FIG. 5F shows the flow of processing of the interrupt from the channels (5-8). Also, FIG. 5G shows the flow of processing of the interrupt from the channels (13-16). The details of the processing are the same as the processing of the interrupts from the channels (1-4) and the channels (9-12) and therefore need not be described.
  • step 50 returns to step 200 ⁇ step 202 ⁇ step 204 ⁇ step 206.
  • step 212 and subsequent steps the operation for image tone decision is effected, and the condition for this is after all of 144 pixel data of the first 9 lines (line 1- line 9) ⁇ (An1-An16) have been stored. That is, after the data of the first 9 lines have been stored, image tone decision is effected each time 16 pixels of 1 line are stored.
  • pixel decision becomes successively possible by using an image processing processor and therefore, it has become unnecessary to effect decision and outputting at a block unit each time 4 ⁇ 4 or 8 ⁇ 8 blocks shift as has heretofore been effected on hard logic. Therefore, with the 8 pixels (encircled by solid line) of (5, 5)-(12, 5) of line 5 of the memory map shown in FIG. 3A as the object, the block of 9 ⁇ 9 pixels centering around each of the 8 pixels is processed in a predetermined manner and binarization is effected by a threshold value suitable for the pixel concerned.
  • the counter, etc. used in the following flow will now be again described by reference to FIGS. 3B and 3C.
  • the range in which image tone recognition is done is the 9 ⁇ 9 matrix of FIG. 3C.
  • the counter w is varied so that this matrix can move in 16 ⁇ 9 memory map.
  • the pixels in the 9 ⁇ 9 matrix are designated by (x+w, y) as shown in FIG. 3C. (w+5, 5) is the central pixel position at which binarization is effected.
  • the counter c is a counter for designating the bit position in RES register.
  • max register and min register are registers for holding the maximum and minimum densities, respectively, in 9 ⁇ 9, and their initial values are 0 and 63, respectively.
  • the max register may assume a value of 0-63.
  • the range thereof should be 0-255, but for the reason that the processing capacity becomes great and that there is little or no influence imparted to the image even if the image is processed by the six most significant bits of 8 bits, the decision is all within the range of 0-63 of the six most significant bits.
  • steps 220-228 are repeated until the counter x assumes 9 (until it goes to the right end of the 9 ⁇ 9 matrix), and the accumulated value of the absolute value
  • the accumulated value, the maximum value and the minimum value of all 9 lines are stored in the registers S, max and min, respectively.
  • the flow repeated till the next steps 236-240 is a flow in which the aforementioned accumulated value a(x) of the density difference in the sub scan direction is added to the register S.
  • the sum of the density differences between adjoining pixels in the main and sub scan directions in 9 ⁇ 9 is stored in the register S.
  • the counters x and y are returned to their initial values, and at step 246, the content of the register S is divided by the register max minus the register min and the quotient is stored in the register S. This quotient should reflect the image tone in 9 ⁇ 9 and the threshold value of binarization may be chosen in conformity with this number.
  • the steps 250-266 of FIG. 5I are the flow in which this threshold value is chosen. The chosen threshold value is stored in a register T.
  • one of matrix BL (block) 1 to matrix BL8 is chosen in conformity with the value of S, as shown in FIGS. 7A-7H.
  • Each BL is 8 ⁇ 8 matrix and a threshold value thereof is chosen in conformity with the value of a counter BLC.
  • the counter BLC may assume a value of 1-64 and therefore, by varying the value of the counter BLC in succession, the central pixel of 9 ⁇ 9 pixels is binarized by an optimum threshold value while continuous image tone recognition of the block of 9 ⁇ 9 pixels is effected.
  • the density value A(w+5, 5) of the central pixel (w+5, 5) and the register T are compared with each other for binarization. If the density of the pixel is greater than the threshold value, the logical Or of the register RES and the counter c is taken at step 270.
  • the function of the counter c is to store the result of the binarization at a predetermined bit position of the 8-bit register RES because there are eight desired pixels.
  • the counter w is counted up by 1 at step 272.
  • 9 ⁇ 9 matrix can be moved in the 16 ⁇ 9 memory map.
  • the loop of steps 274-278 is a flow for counting up the counter BLC in succession.
  • step 280 the content of the counter c is shifted leftwardly by 1 bit. Accordingly, the value of C becomes one of 1, 2, 4, 8, 16, 32, 64 and 128.
  • the result of the binarization can be stored in the register RES with the bit position shifted in succession.
  • step 282 whether the content of the counter w has become equal to 8 is examined, and step 220 and subsequent steps are repeated until all of the 8 central pixels are binarized, and the result of the binarization is stored in the register RES.
  • step 290 and 292 initialization of the counters x and w is effected. Further, at step 294, the density difference
  • the interior of the memory map is shifted in succession. Accordingly, the ninth line in the memory map becomes empty and the storage of the data by the scanning of the next line becomes possible. Then, at step 316, the flow returns to the return address to wait for the next external interrupt.
  • step 88 of FIG. 5D, step 118 of FIG. 5E, step 148 of FIG. 5F and step 178 of FIG. 5G the flow proceeds to step 92, step 124, step 152 and step 182, respectively, and the pixel densities stored in the CR register are successively stored into the ninth line of the memory map.
  • step 92, step 124, step 152 and step 182 respectively, and the pixel densities stored in the CR register are successively stored into the ninth line of the memory map.
  • the output from the register RES is temporally held in the shift register SR of FIG. 1 and is successively serially put out from the concatenated shift register SR.
  • the parallel-processed image signal is subjected to image tone recognition and binarized on real time.
  • the above-described image tone recognition method is a method whereby image tone recognition is effected while the central pixel is shifted by 1 pixel each, but of course, the same effect may also be obtained by dividing the 16 ⁇ 9 memory map into blocks of the size of a threshold matrix and binarizing each block by the threshold matrix. Again in this case, the image signal in charge of the adjacent CPU is introduced and therefore, non-conformity between the CPUs does not occur and higher speed image processing becomes possible.
  • BL1 (FIG. 7A): This is a dot concentration type matrix in which importance is attached to the color tone representation.
  • the range of the threshold value is 32 tones of 2-62 and the output form is 2 dots/8 ⁇ 8.
  • BL2 (FIG. 7B): This is a dot concentration type matrix in which importance is attached to the color tone representation and resolution is also value is 32 tones of 2-62 and the output form is 4 dots/18 ⁇ 8.
  • BL3 (FIG. 7C): This is a dither processing matrix in which resolution is taken into consideration and contrast is emphasized.
  • the range of the threshold value is 17 tones of 16-48 and the output form is 4 dots/8 ⁇ 8.
  • BL4 (FIG. 7D): This is a matrix by a predetermined threshold value (32). 2 tones are provided by a binarizing process.
  • BL5 (FIG. 7E): This is a dither processing matrix in which, as in BL3, resolution is taken into consideration and contrast is emphasized.
  • the range of the threshold value is 17 tones of 16-48 and the output form is 4 dots/8 ⁇ 8.
  • BL6 (FIG. 7F): This is a dither processing matrix comprising a small matrix which is the object of a net-point original.
  • the range of the threshold value is 5 tones of 20, 28, 36 and 44 and the output form is 16 dots/8 ⁇ 8.
  • BL7 (FIG. 7G): This is a matrix in which the phase of BL6 is changed by 90° with a view to suppress moire.
  • BL8 (FIG. 7H): This is a matrix in which the phase of BL6 is changed by 180° with a view to suppress moire.
  • FIGS. 8 and 9 show another embodiment of the CPU.
  • the image signal from the adjacent CPU has been introduced into the A/D converting portion, but in this embodiment, the pixel density A/D-converted by the adjacent CPU is introduced through external buses B0-B15. By doing so, there is also obtained the memory map as shown in FIG. 10.
  • a plurality of image processing means effect image processing in parallel, whereby high-speed image processing becomes possible.
  • input means such as sensor cells and image processing means can be arranged on the same base plate and thus, compactness of the apparatus is achieved.
  • the respective image processing means each having an A/D converting portion therein start A/D conversion and effect image processing substantially simultaneously with the sample signal obtained in synchronism with the main scanning and therefore, high-speed two-dimensional image processing becomes possible.
  • individual image processing means receive as an input a part of the image signal in charge of the adjacent image processing means, whereby non-conformity of image processing does not occur.

Landscapes

  • Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Signal Processing (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Facsimile Scanning Arrangements (AREA)
  • Facsimile Image Signal Circuits (AREA)
  • Image Processing (AREA)
US07/147,091 1984-12-14 1988-01-20 Apparatus for parallel-processing image data Expired - Lifetime US4786820A (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
JP59262888A JPS61141264A (ja) 1984-12-14 1984-12-14 画像処理装置
JP59-262888 1984-12-14

Related Parent Applications (1)

Application Number Title Priority Date Filing Date
US06807662 Continuation 1985-12-11

Publications (1)

Publication Number Publication Date
US4786820A true US4786820A (en) 1988-11-22

Family

ID=17382010

Family Applications (1)

Application Number Title Priority Date Filing Date
US07/147,091 Expired - Lifetime US4786820A (en) 1984-12-14 1988-01-20 Apparatus for parallel-processing image data

Country Status (2)

Country Link
US (1) US4786820A (enrdf_load_stackoverflow)
JP (1) JPS61141264A (enrdf_load_stackoverflow)

Cited By (26)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5047931A (en) * 1989-05-30 1991-09-10 General Electric Company Parallel computation of fan-beam back-projection reconstruction algorithm in computed tomography
US5128760A (en) * 1990-02-28 1992-07-07 Texas Instruments Incorporated Television scan rate converter
EP0535966A3 (en) * 1991-10-03 1993-08-11 Ncr International Inc. Image processing method and system
US5293481A (en) * 1987-02-18 1994-03-08 Canon Kabushiki Kaisha Data parallel processing apparatus
US5450503A (en) * 1985-07-04 1995-09-12 Canon Kabushiki Kaisha Image recognition method
US5463732A (en) * 1994-05-13 1995-10-31 David Sarnoff Research Center, Inc. Method and apparatus for accessing a distributed data buffer
US5701505A (en) * 1992-09-14 1997-12-23 Fuji Xerox Co., Ltd. Image data parallel processing apparatus
EP0535948B1 (en) * 1991-10-03 1998-09-09 NCR International, Inc. Method and system for compressing data
DE3844921C2 (de) * 1987-02-18 1999-01-21 Canon Kk Bildverarbeitungsgerät
US5892546A (en) * 1991-06-25 1999-04-06 Canon Kabushiki Kaisha Encoding apparatus and method for encoding a quantized difference between an input signal and a prediction value
EP0902585A3 (en) * 1997-09-11 1999-12-08 Sharp Kabushiki Kaisha Method and apparatus for image processing
DE3845004C2 (de) * 1987-02-18 2002-04-04 Canon Kk Bildverarbeitungseinrichtung
US6477281B2 (en) 1987-02-18 2002-11-05 Canon Kabushiki Kaisha Image processing system having multiple processors for performing parallel image data processing
EP0967792A3 (en) * 1998-06-26 2004-06-16 Sony Corporation Printer having image correcting capability
US20070002408A1 (en) * 2005-06-30 2007-01-04 Brother Kogyo Kabushiki Kaisha Image-reading device
US20090128120A1 (en) * 2007-11-15 2009-05-21 Seiko Epson Corporation Reference voltage generation circuit, ad converter, da converter, and image processor
US20130024660A1 (en) * 1997-07-15 2013-01-24 Kia Silverbrook Portable handheld device with multi-core image processor
US8789939B2 (en) 1998-11-09 2014-07-29 Google Inc. Print media cartridge with ink supply manifold
US8823823B2 (en) 1997-07-15 2014-09-02 Google Inc. Portable imaging device with multi-core processor and orientation sensor
US8866923B2 (en) 1999-05-25 2014-10-21 Google Inc. Modular camera and printer
US8896724B2 (en) 1997-07-15 2014-11-25 Google Inc. Camera system to facilitate a cascade of imaging effects
US8902333B2 (en) 1997-07-15 2014-12-02 Google Inc. Image processing method using sensed eye position
US8908075B2 (en) 1997-07-15 2014-12-09 Google Inc. Image capture and processing integrated circuit for a camera
US8936196B2 (en) 1997-07-15 2015-01-20 Google Inc. Camera unit incorporating program script scanner
US9055221B2 (en) 1997-07-15 2015-06-09 Google Inc. Portable hand-held device for deblurring sensed images
EP2161685A3 (en) * 2008-09-09 2016-11-23 Sony Corporation Pipelined image processing engine

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US3995107A (en) * 1974-05-08 1976-11-30 Rca Corporation Charge coupled parallel-to-serial converter for scene scanning and display
US4222105A (en) * 1977-03-28 1980-09-09 Canon Kabushiki Kaisha Recording apparatus including analog-digital converter
US4513313A (en) * 1982-12-07 1985-04-23 Canon Kabushiki Kaisha Solid state imaging device
US4584609A (en) * 1985-04-15 1986-04-22 Rca Corporation Parallel-input/serial output CCD register with clocking noise cancellation, as for use in solid-state imagers
US4589034A (en) * 1980-12-05 1986-05-13 Canon Kabushiki Kaisha Image processing apparatus

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US3995107A (en) * 1974-05-08 1976-11-30 Rca Corporation Charge coupled parallel-to-serial converter for scene scanning and display
US4222105A (en) * 1977-03-28 1980-09-09 Canon Kabushiki Kaisha Recording apparatus including analog-digital converter
US4589034A (en) * 1980-12-05 1986-05-13 Canon Kabushiki Kaisha Image processing apparatus
US4513313A (en) * 1982-12-07 1985-04-23 Canon Kabushiki Kaisha Solid state imaging device
US4584609A (en) * 1985-04-15 1986-04-22 Rca Corporation Parallel-input/serial output CCD register with clocking noise cancellation, as for use in solid-state imagers

Cited By (84)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5450503A (en) * 1985-07-04 1995-09-12 Canon Kabushiki Kaisha Image recognition method
DE3844921C2 (de) * 1987-02-18 1999-01-21 Canon Kk Bildverarbeitungsgerät
DE3845004C2 (de) * 1987-02-18 2002-04-04 Canon Kk Bildverarbeitungseinrichtung
US5293481A (en) * 1987-02-18 1994-03-08 Canon Kabushiki Kaisha Data parallel processing apparatus
US6477281B2 (en) 1987-02-18 2002-11-05 Canon Kabushiki Kaisha Image processing system having multiple processors for performing parallel image data processing
DE3845005C2 (de) * 1987-02-18 2002-04-04 Canon Kk Bildverarbeitungseinrichtung
US5047931A (en) * 1989-05-30 1991-09-10 General Electric Company Parallel computation of fan-beam back-projection reconstruction algorithm in computed tomography
US5128760A (en) * 1990-02-28 1992-07-07 Texas Instruments Incorporated Television scan rate converter
US5900910A (en) * 1991-06-25 1999-05-04 Canon Kabushiki Kaisha Block matching method and apparatus which can use luminance/chrominance data
US5892546A (en) * 1991-06-25 1999-04-06 Canon Kabushiki Kaisha Encoding apparatus and method for encoding a quantized difference between an input signal and a prediction value
EP0535966A3 (en) * 1991-10-03 1993-08-11 Ncr International Inc. Image processing method and system
EP0535948B1 (en) * 1991-10-03 1998-09-09 NCR International, Inc. Method and system for compressing data
US5701505A (en) * 1992-09-14 1997-12-23 Fuji Xerox Co., Ltd. Image data parallel processing apparatus
US5463732A (en) * 1994-05-13 1995-10-31 David Sarnoff Research Center, Inc. Method and apparatus for accessing a distributed data buffer
US9544451B2 (en) 1997-07-12 2017-01-10 Google Inc. Multi-core image processor for portable device
US9338312B2 (en) 1997-07-12 2016-05-10 Google Inc. Portable handheld device with multi-core image processor
US8902340B2 (en) 1997-07-12 2014-12-02 Google Inc. Multi-core image processor for portable device
US8947592B2 (en) 1997-07-12 2015-02-03 Google Inc. Handheld imaging device with image processor provided with multiple parallel processing units
US9131083B2 (en) 1997-07-15 2015-09-08 Google Inc. Portable imaging device with multi-core processor
US8934027B2 (en) 1997-07-15 2015-01-13 Google Inc. Portable device with image sensors and multi-core processor
US9584681B2 (en) 1997-07-15 2017-02-28 Google Inc. Handheld imaging device incorporating multi-core image processor
US9560221B2 (en) 1997-07-15 2017-01-31 Google Inc. Handheld imaging device with VLIW image processor
US9432529B2 (en) 1997-07-15 2016-08-30 Google Inc. Portable handheld device with multi-core microcoded image processor
US9237244B2 (en) 1997-07-15 2016-01-12 Google Inc. Handheld digital camera device with orientation sensing and decoding capabilities
US9219832B2 (en) * 1997-07-15 2015-12-22 Google Inc. Portable handheld device with multi-core image processor
US9197767B2 (en) 1997-07-15 2015-11-24 Google Inc. Digital camera having image processor and printer
US9191530B2 (en) 1997-07-15 2015-11-17 Google Inc. Portable hand-held device having quad core image processor
US9191529B2 (en) 1997-07-15 2015-11-17 Google Inc Quad-core camera processor
US20130024660A1 (en) * 1997-07-15 2013-01-24 Kia Silverbrook Portable handheld device with multi-core image processor
US9185246B2 (en) 1997-07-15 2015-11-10 Google Inc. Camera system comprising color display and processor for decoding data blocks in printed coding pattern
US9185247B2 (en) 1997-07-15 2015-11-10 Google Inc. Central processor with multiple programmable processor units
US8823823B2 (en) 1997-07-15 2014-09-02 Google Inc. Portable imaging device with multi-core processor and orientation sensor
US8836809B2 (en) 1997-07-15 2014-09-16 Google Inc. Quad-core image processor for facial detection
US8866926B2 (en) 1997-07-15 2014-10-21 Google Inc. Multi-core processor for hand-held, image capture device
US9179020B2 (en) 1997-07-15 2015-11-03 Google Inc. Handheld imaging device with integrated chip incorporating on shared wafer image processor and central processor
US8896720B2 (en) 1997-07-15 2014-11-25 Google Inc. Hand held image capture device with multi-core processor for facial detection
US8896724B2 (en) 1997-07-15 2014-11-25 Google Inc. Camera system to facilitate a cascade of imaging effects
US8902357B2 (en) 1997-07-15 2014-12-02 Google Inc. Quad-core image processor
US9168761B2 (en) 1997-07-15 2015-10-27 Google Inc. Disposable digital camera with printing assembly
US8902324B2 (en) 1997-07-15 2014-12-02 Google Inc. Quad-core image processor for device with image display
US8902333B2 (en) 1997-07-15 2014-12-02 Google Inc. Image processing method using sensed eye position
US8908075B2 (en) 1997-07-15 2014-12-09 Google Inc. Image capture and processing integrated circuit for a camera
US8908051B2 (en) 1997-07-15 2014-12-09 Google Inc. Handheld imaging device with system-on-chip microcontroller incorporating on shared wafer image processor and image sensor
US8908069B2 (en) 1997-07-15 2014-12-09 Google Inc. Handheld imaging device with quad-core image processor integrating image sensor interface
US8913151B2 (en) 1997-07-15 2014-12-16 Google Inc. Digital camera with quad core processor
US8913182B2 (en) 1997-07-15 2014-12-16 Google Inc. Portable hand-held device having networked quad core processor
US8913137B2 (en) 1997-07-15 2014-12-16 Google Inc. Handheld imaging device with multi-core image processor integrating image sensor interface
US8922670B2 (en) 1997-07-15 2014-12-30 Google Inc. Portable hand-held device having stereoscopic image camera
US8922791B2 (en) 1997-07-15 2014-12-30 Google Inc. Camera system with color display and processor for Reed-Solomon decoding
US8928897B2 (en) 1997-07-15 2015-01-06 Google Inc. Portable handheld device with multi-core image processor
US8934053B2 (en) 1997-07-15 2015-01-13 Google Inc. Hand-held quad core processing apparatus
US9148530B2 (en) 1997-07-15 2015-09-29 Google Inc. Handheld imaging device with multi-core image processor integrating common bus interface and dedicated image sensor interface
US8937727B2 (en) 1997-07-15 2015-01-20 Google Inc. Portable handheld device with multi-core image processor
US8936196B2 (en) 1997-07-15 2015-01-20 Google Inc. Camera unit incorporating program script scanner
US8947679B2 (en) 1997-07-15 2015-02-03 Google Inc. Portable handheld device with multi-core microcoded image processor
US9143636B2 (en) 1997-07-15 2015-09-22 Google Inc. Portable device with dual image sensors and quad-core processor
US8953178B2 (en) 1997-07-15 2015-02-10 Google Inc. Camera system with color display and processor for reed-solomon decoding
US8953061B2 (en) 1997-07-15 2015-02-10 Google Inc. Image capture device with linked multi-core processor and orientation sensor
US8953060B2 (en) 1997-07-15 2015-02-10 Google Inc. Hand held image capture device with multi-core processor and wireless interface to input device
US9055221B2 (en) 1997-07-15 2015-06-09 Google Inc. Portable hand-held device for deblurring sensed images
US9060128B2 (en) 1997-07-15 2015-06-16 Google Inc. Portable hand-held device for manipulating images
US9124736B2 (en) 1997-07-15 2015-09-01 Google Inc. Portable hand-held device for displaying oriented images
US9124737B2 (en) 1997-07-15 2015-09-01 Google Inc. Portable device with image sensor and quad-core processor for multi-point focus image capture
US9143635B2 (en) 1997-07-15 2015-09-22 Google Inc. Camera with linked parallel processor cores
US9137398B2 (en) 1997-07-15 2015-09-15 Google Inc. Multi-core processor for portable device with dual image sensors
US9137397B2 (en) 1997-07-15 2015-09-15 Google Inc. Image sensing and printing device
US6111982A (en) * 1997-09-11 2000-08-29 Sharp Kabushiki Kaisha Image processing apparatus and recording medium recording a program for image processing
EP0902585A3 (en) * 1997-09-11 1999-12-08 Sharp Kabushiki Kaisha Method and apparatus for image processing
US7202975B2 (en) 1998-06-26 2007-04-10 Sony Corporation Printer having image correcting capability
US20050030567A1 (en) * 1998-06-26 2005-02-10 Sony Corporation Printer having image correcting capability
US20050041260A1 (en) * 1998-06-26 2005-02-24 Sony Corporation Printer having image correcting capability
US20050041261A1 (en) * 1998-06-26 2005-02-24 Sony Corporation Printer having image correcting capability
US6831755B1 (en) 1998-06-26 2004-12-14 Sony Corporation Printer having image correcting capability
US20050030568A1 (en) * 1998-06-26 2005-02-10 Sony Corporation. Printer having image correcting capability
EP0967792A3 (en) * 1998-06-26 2004-06-16 Sony Corporation Printer having image correcting capability
US7199904B2 (en) 1998-06-26 2007-04-03 Sony Corporation Printer having image correcting capability
US7196818B2 (en) 1998-06-26 2007-03-27 Sony Corporation Printer having image correcting capability
US7193749B2 (en) 1998-06-26 2007-03-20 Sony Corporation Printer having image correcting capability
US8789939B2 (en) 1998-11-09 2014-07-29 Google Inc. Print media cartridge with ink supply manifold
US8866923B2 (en) 1999-05-25 2014-10-21 Google Inc. Modular camera and printer
US20070002408A1 (en) * 2005-06-30 2007-01-04 Brother Kogyo Kabushiki Kaisha Image-reading device
US8693063B2 (en) * 2005-06-30 2014-04-08 Brother Kogyo Kabushiki Kaisha Image-reading device
US20090128120A1 (en) * 2007-11-15 2009-05-21 Seiko Epson Corporation Reference voltage generation circuit, ad converter, da converter, and image processor
EP2161685A3 (en) * 2008-09-09 2016-11-23 Sony Corporation Pipelined image processing engine

Also Published As

Publication number Publication date
JPH0558310B2 (enrdf_load_stackoverflow) 1993-08-26
JPS61141264A (ja) 1986-06-28

Similar Documents

Publication Publication Date Title
US4786820A (en) Apparatus for parallel-processing image data
EP0238976B1 (en) Method of tint generation and apparatus therefor
EP0041400B1 (en) Multi-resolution image signal processing apparatus and method
US4646355A (en) Method and apparatus for input picture enhancement by removal of undersired dots and voids
US4291334A (en) System for detecting the position of an object
JPS6364114B2 (enrdf_load_stackoverflow)
CA1066806A (en) Electronic system for reading symbols
JPH04185074A (ja) 画像の解像度変換方法
US3952290A (en) Read-only optical memory system
EP0273638A2 (en) Image scanner systems
US4799154A (en) Array processor apparatus
CA2271868A1 (en) Pipeline processor for medical and biological image analysis
JPH0556062B2 (enrdf_load_stackoverflow)
JPH0556063B2 (enrdf_load_stackoverflow)
JPH1093819A (ja) 画像処理装置
JPH0376377A (ja) 画像処理装置
EP0418844A2 (en) Color image information processing method
JP2861089B2 (ja) 画線付加装置
JPH0896116A (ja) 画像信号処理装置及び方法
EP0428624A4 (en) Intelligent scan image processor
JP3061830B2 (ja) 画像処理装置
JP2861845B2 (ja) 白抜き機能付きイメージスキャナ
JPH09321989A (ja) エッジ領域検出装置
Chang et al. Programmable accelerating chip for optical Chinese character recognition
JP2004289846A (ja) 取込画像のノイズを補償するシステムおよび方法

Legal Events

Date Code Title Description
STCF Information on status: patent grant

Free format text: PATENTED CASE

CC Certificate of correction
FPAY Fee payment

Year of fee payment: 4

FPAY Fee payment

Year of fee payment: 8

FEPP Fee payment procedure

Free format text: PAYOR NUMBER ASSIGNED (ORIGINAL EVENT CODE: ASPN); ENTITY STATUS OF PATENT OWNER: LARGE ENTITY

Free format text: PAYER NUMBER DE-ASSIGNED (ORIGINAL EVENT CODE: RMPN); ENTITY STATUS OF PATENT OWNER: LARGE ENTITY

FPAY Fee payment

Year of fee payment: 12